1. Field of the Invention
The present invention relates to an apparatus for testing body fluid constituents. More particularly, the present invention pertains to an apparatus for collecting body fluid for testing.
2. Description of the Art
The prior art has long been seeking procedures for testing and determining the level of blood constituents. Particularly, a great deal of attention has been spent on the development of techniques for measuring blood glucose.
Historically, blood glucose and other bodily analyte measurements were, and remain, invasive. Such measurements are generally made by withdrawing a blood sample and measuring the desired analyte within the blood or plasma. Blood samples can be withdrawn by inserting a needle into a major artery or, more commonly, a vein. A syringe or other device is used to provide any necessary suction and collect the blood sample. Needles used for this sampling technique must be long enough to pass through the skin, subcutaneous tissue, and blood vessel wall. The needle must also have a sufficient diameter to allow timely collection of the blood sample without causing hemolysis of the blood. Minimal diameter to meet these criteria is generally 20 gauge or larger diameter. Such direct vascular blood sampling has several limitations, including pain, hematoma and other bleeding complications, and infection. In addition, due to the vascular damage resulting from the needle puncture, sampling could not be repeated on a routine basis. Finally, it is extremely difficult for patients to perform a direct vascular puncture on themselves.
The other common technique for collecting a blood sample is to cut or lance the skin and the subcutaneous tissue, including the small, underlying blood vessels, to produce a localized bleeding on the body surface. A lancet, knife, or other cutting device is required. The blood on the body surface can then be collected into a small tube or other container. The fingertip is the most frequently used site to collect blood in this method due to the large number of small blood vessels located in the region. One method is shown in U.S. Pat. 4,637,403. This sampling method also suffers from several major disadvantages, including pain and the potential for infection and other problems associated with repeated sampling for a confined area. Pain is a major disadvantage since the fingertip has a large concentration of nerve endings. Also, there is a limited body surface area from which to take these samples and measurement on a high frequency basis.
Because the prior art invasive techniques are painful, patients frequently avoid having blood glucose measured. For diabetics, the failure to measure blood glucose on a prescribed basis can be very dangerous. Also, the invasive techniques, which would result in lancing blood vessels, create an enhanced risk for disease transmission.
Attempts have been made to develop glucose and other analyte sensors for implantation in the human body. Implanted glucose sensors would be primarily to control insulin infusion pumps or provide continuous, chronic monitoring. Development of a permanently implanted or long-term, chronic implanted sensor has been unsuccessful. Attempts to develop short-term implantable sensors (up to 2-3 days) have also met with very limited success. Most implantable sensors are based on measuring various products from chemical reactions between agent(s) located on or within the sensor and the desired analyte. Implanted glucose sensors have typically used the glucose oxidase reaction to measure the amount of glucose, as described in U.S. Pat. No. 5,108,819. Such implantable glucose sensors have been intended for insertion through the epidermis and dermis to the subcutaneous tissue. An alternative location previously described for chronic sensor implant is the peritoneal cavity. All such implanted sensors require direct or telemetered connection to a measurement instrument, usually located external the body.
All implanted sensors are faced with several major problems. First, all foreign materials, including materials incorporated into a glucose sensor, produce unwanted body reactions. Such reactions include the formation of fibrotic tissue around the sensor which alters the sensor's contact with normal body fluids and analytes, such as glucose. The body's natural defense mechanism may also have a direct "poisoning" effect upon the sensor's operation by interfering with the chemical reactions required by chemical-based sensors. As with any implanted object, implanted sensors may also initiate other bodily reactions including inflammation, pain, tissue necrosis, infection, and other unwanted reactions.
Implanted sensors require certain chemicals and chemical reactions to determine the level of analyte in the surrounding medium. These chemical reactions are the source of the other major problem facing any implantable sensor. Chemically-based sensors require products to be consumed and other products to be produced as part of the sensor's normal operations. Therefore, the sensors can quickly be depleted of the chemical agents required to sustain the desired chemical reactions. Secondly, by-products are given off as a result of the basic chemical reaction. These by-products often "poison" the sensor or cause other unwanted tissue reactivity. Because of these severe limitations, implanted sensors are not practical. Finally, such implanted sensors are painful to implant and are a source of infection.
By withdrawing the body fluid containing the glucose or other analyte and making the measurement outside the body, these aforementioned sensor based problems can be avoided. Specifically, there is no concern about the chronic tissue response to the foreign sensor material or the limited operational life of the sensor due to the consumption of reaction agents or the production of unwanted by-products from that reaction.
In view of the risk associated with invasive techniques, the prior art has sought to develop non-invasive blood glucose measurement techniques. An example of such is shown in U.S. Pat. No. 4,882,492 to Schlager. Schlager teaches a non-invasive near-infrared measurement of blood. Schlager is particularly directed to the measurement of blood glucose levels. The Schlager patent recognizes that certain wavelengths of light in the near-infrared spectrum are absorbed by glucose. Modulated light is directed against a tissue (shown as an earlobe). The light is either passed through the tissue or impinged on a skin surface. The light is spectrally modified in response to the amount of analyte (for example, glucose) in the blood and tissue. The spectrally modified light is split with one beam passed through a correlation cell. The other beam is passed through a reference cell. The intensity of the beams passing through the correlation cell and the reference cell are compared to calculate a glucose concentration in the sample. Other non-invasive blood glucose methods are shown in U.S. Pat. No. 4,805,623, 4,655,225, 4,014,321 and 3,958,560.
One drawback of prior art non-invasive systems is that by passing the infrared light through a complex medium (such as an earlobe) very complex data is generated. Algorithms must be developed to manipulate the data in order to attempt to provide reliable indications of blood glucose measurements. Also, such devices may require exact placement of the measuring device (e.g., precise placement on a patient's finger or near an earlobe) to minimize measurement error. Such devices may also be difficult to calibrate. To date, the prior art has not developed commercially available non-invasive methods which provide accurate data.
In addition to the foregoing, applicants' assignee is the owner of various patents pertaining to blood glucose measurement. For example, U.S. Pat. No. 5,179,951 to Knudson dated Jan. 19, 1993 teaches an invasive blood glucose measurement where infrared light is passed through a sample of blood by use of an implanted catheter. Similarly, U.S. Pat. No. 5,079,421 teaches such a system.
U.S. Pat. No. 5,146,091 teaches a non-invasive blood glucose measurement utilizing FTIR (Fourier Transform Infrared) techniques to determine blood glucose levels and U.S. Pat. No. 5,115,133 which directs infrared light to the eardrum. As indicated in the aforementioned commonly assigned patents, the testing wavelength includes a glucose sensitive wavelength of about 500 to about 4,000 wave numbers (cm.sup.-1). Preferably, the glucose absorbable wavelength is about 1,040 wave numbers.
It is an object of the present invention to provide an enhanced technique for collecting a sample fluid and for measuring fluid constituents in the sample.